DiggTop

Llama-4-Scout-17B-16E-Instruct

Llama family · 109.0B params

1recorded run
44.5best tok/s
1verified
Fastest GPUs for this model (best tok/s)
  1. CMP 170HX44.5

All benchmarks Methodology

GPUModelQuantBackendCtx tok/sprompt tok/sVRAM GBSource
CMP 170HX · 64 GBLlama-4-Scout-17B-16E-InstructUD-Q3_K_XLllama.cpp 0.28.0204844.5 584.446.6src